Connect securely or upload your data  ·  No ads  ·  We never sell your financial data

Method · the WealthAge Resilience Score

How the WealthAge Resilience Score is validated

The WealthAge Resilience Score™ is a personal 0–1000 measure of how prepared your current financial picture is for a financial shock. It works like a credit score, one number you own and watch, but it answers your question, not a lender's: if the income stopped, how long would you stand? Instead of averaging opinions, it simulates a shock against your real finances (what you could actually access, what keeps draining, what income would replace) across six determinants: liquid runway, income adequacy, committed costs, debt service, income replacement, and protective context. It was validated on the U.S. Census Bureau's Survey of Income and Program Participation, where it predicted real hardship better than savings-runway alone. We publish the method and show your reason codes; we don't publish the machinery. It's informational and yours: not a credit score, consumer report, lending input, or guarantee.

By Steven Sterling, who built it: the score, the engine, and the validation study.

WealthAge Resilience Score · specimen

Score687 of 1000
Runway5.2 months
Holding it upliquid runway
Holding it downincome replacement

illustrative fixture, not a real user · every score ships with its reason codes, exactly like this

The engine · seven steps

What the simulation does

The engine doesn't grade your habits. It runs the shock, step by step, the same way for every household.

01 Route the household. Earning income, or living on passive income. A retiree is never scored on a missing paycheck.
02 Compute the accessible buffer. Every asset counted by how reachable it really is in a crisis: cash fully; brokerage, retirement, home equity, and crypto progressively less, because "liquid-looking" assets fall in the same downturn that costs you the job.
03 Compute committed monthly outflow. Housing, utilities, a food floor, minimum debt, insurance. Discretionary spending is excluded: it can be cut, so it isn't fragility.
04 Compute replacement income during the shock. Other earners, passive income, time-limited unemployment insurance, a capped family draw. AI-occupation exposure lengthens the expected time back to work.
05 Effective runway. Buffer ÷ (committed outflow − replacement): the honest "months you'd survive."
06 Combine six determinants into the 0–1000 score. Liquid runway weighs heaviest; protective context lightest. About 800 reads as fully resilient; 950+ is rare.
07 Severity gates cap severe fragility. Insolvency, sub-month runway, or crushing debt service caps the score. Strengths can never average away a fatal flaw. That's what makes it a judgment instrument, not a spreadsheet average.

What the score weighs

The six determinants

Liquid runway (weighs heaviest) · income adequacy · committed costs · debt service · income replacement (including AI-occupation exposure) · protective context (weighs lightest). Whenever your score is shown, you see each determinant's direction and which weighs heaviest and lightest for you. Your reason codes, always.

Liquid runway weighs heaviest 62
Income adequacy 74
Committed costs 58
Debt service 81
Income replacement incl. AI exposure 55
Protective context weighs lightest 70

Illustrative fixture sub-scores, not weights and not a real user, shown to make the ordering (heaviest to lightest) visible instead of only stated in prose.

The evidence

How we know it works

The model was validated on the U.S. Census Bureau's Survey of Income and Program Participation (SIPP): 24,162 real households: their 2022 finances against whether they actually hit hardship in 2023. Given two households, one of which hits hardship, the months-of-expenses rule picks the right one 64% of the time; the simulation picks it 72% of the time. In formal terms, AUC 0.717 against 0.640. The protocol, for the skeptical reader: the model was fit and evaluated on held-out households it never saw during calibration, and AUC measures exactly the pairwise comparison above, so 0.5 is a coin flip and 1.0 is omniscience. 0.717 beats the rule of thumb's 0.640 on the same held-out data. Better than the rule of thumb, and honestly short of an oracle, which is why every score carries a confidence read based on how complete your data is. Income is not resilience: in our fixture gallery, an $18k-a-month over-committed earner scores 422 while a modest, disciplined saver scores 814. Every example number on this site (687, 5.2 months, 422, 814) is an illustrative fixture, never a real user.

AUC measures how often the instrument ranks a household that later hit hardship as more fragile than one that didn't. 1.0 would be an oracle; 0.5 is chance.

Fixture personas from the published gallery, not real users.

The boundary

What we publish, and what we don't

We publish the method what goes in, why, and how well it predicts.
We don't publish the machinery the exact weights, thresholds, and calibration.

That is the same boundary every serious score keeps.

The disclaimers, plainly

What the score is not

Not a credit score and not a consumer report or lending input.
Not a guarantee and never a verdict on your past.
Not a game no tiers, no grades, no badges.

It describes your position; what you do with it is yours.

A living number

How the score stays current

Your score recomputes from your latest successfully synced or uploaded data, and every score surface carries its as-of time. Structural facts like occupation, insurance, and dependents change on life events, not daily. A separate, dated monthly Macro Climate read on the world is shown beside your number, never blended into it: when a scenario reading moves on macro alone, your finances didn't change. The world did.

Get WealthAge

$129 a year, or $14.99 a month

See how it works →